A Better Match for Drivers and Riders: Reinforcement Learning at Lyft-Reference-Cited by-同舟云学术

登录注册会员服务联系我们

A Better Match for Drivers and Riders: Reinforcement Learning at Lyft

Published:2024-01 Issue:1 Volume:54 Page:71-83
ISSN:2644-0865
Container-title:INFORMS Journal on Applied Analytics
language:en
Short-container-title:INFORMS Journal on Applied Analytics

Author:

Azagirre Xabi¹,Balwally Akshay¹,Candeli Guillaume¹,Chamandy Nicholas¹,Han Benjamin¹,King Alona¹,Lee Hyungjun¹,Loncaric Martin¹,Martin Sébastien²^ORCID,Narasiman Vijay¹,Qin Zhiwei (Tony)¹,Richard Baptiste¹,Smoot Sara¹,Taylor Sean¹,van Ryzin Garrett¹,Wu Di¹,Yu Fei¹,Zamoshchin Alex¹

Affiliation:

1. Lyft, San Francisco, California 94107;

2. Northwestern University, Evanston, Illinois 60208

Abstract

We used reinforcement learning to improve how Lyft matches drivers and riders. The change was implemented globally and led to more than $30 million per year in incremental driver revenue.

Publisher

Institute for Operations Research and the Management Sciences (INFORMS)

Link

https://pubsonline.informs.org/doi/pdf/10.1287/inte.2023.0083

Reference14 articles.

1. Online Vehicle Routing: The Edge of Optimization in Large-Scale Applications

2. Parking infrastructure: energy, emissions, and automobile life-cycle environmental accounting

3. Dynamic Type Matching

4. The Hungarian method for the assignment problem

5. Online spatio-temporal matching in stochastic and dynamic domains

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370

www.globalauthorid.com

TOP